2 The procedure
2.1.1 Lipoatrophy is the localised loss of fat from within subcutaneous tissue. It can be a congenital condition or can be associated with subcutaneous injection sites. Facial lipoatrophy is commonly seen after HIV treatment, particularly with the older antiretroviral drugs. It involves wasting of the soft tissues of the cheeks, temples and around the eyes, producing changes in appearance that have severe psychological and social consequences for some patients.
2.1.2 Current treatments for HIV-associated lipoatrophy include autologous fat transfer, dermal fat grafting, transfer of skin flaps and injection of temporary dermal fillers (such as collagen) or semi-permanent dermal fillers (such as polylactic acid).
2.2.1 Deep dermal injection of permanent or non-absorbable gel polymer aims to improve the appearance of HIV-associated lipoatrophy and its related psychological effects. It is intended to achieve a long-lasting result.
2.2.2 The procedure is performed under general or local anaesthesia. Non-absorbable gel polymer is injected with a needle or cannula, deep into the subcutaneous tissue. Strict aseptic technique is used and prophylactic antibiotics are given. After injection the gel is massaged into position to produce a good aesthetic result. Once in place, the gel forms a thin external membrane or capsule that isolates it from the surrounding tissues and results in a liquid-filled endoprosthesis. The volume of gel injected depends on which part of the body is being treated and the degree of lipoatrophy; for facial treatment typically 1 ml is injected at each site at each treatment session. A course of injections over several weeks may be needed depending on the volume required.
Sections 2.3 and 2.4 describe efficacy and safety outcomes from the published literature that the Committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see the overview.
2.3.1 A non-randomised comparative study of 299 patients (130 treated by NAGP injection, 91 by polylactic acid [PLA] injection, 54 by autologous fat transfer only [AFT] and 24 by AFT plus PLA injection) reported significant improvements in facial aesthetic satisfaction (p<0.0001), body image satisfaction (p<0.0001) and depression score (p=0.014) at 48-week follow-up for the NAGP group. Patients were evaluated by visual analogue scale, the Assessment of Body Change and Distress questionnaire and the Beck Depression Inventory scale. A cohort study of 32 patients (5 from a pilot study and 27 from a randomised controlled trial comparing immediate against delayed dermal injection of NAGP) reported significant improvements in scores for anxiety (p<0.001) and depression (p<0.001) on the Hospital Anxiety Depression scale, the slightly modified Dermatology Quality of Life Survey (p<0.001), and the mental health domain (p=0.02) of the Medical Outcomes Study-HIV health survey at 4-year follow-up compared with baseline.
2.3.2 The cohort study of 32 patients reported significant improvements in scores for median physician and patient-graded facial lipoatrophy severity (−2 [interquartile range −2, −1; p<0.001] and −1 [interquartile range −3, −1; p<0.001]) at 4-year follow-up compared with baseline.
2.3.3 A case series of 145 patients reported that 89% of patients were 'satisfied' or 'very satisfied' with the results 4 years after receiving NAGP injections (assessed using a 3-point scale ranging from not satisfied to very satisfied).
2.3.4 The non-randomised comparative study of 299 patients reported a significant augmentation of cheek thickness (right cheek from 4.3±1.9 mm to 9.5 mm p<0.0001, left cheek from 4.4±2 mm to 9.6±3.1 mm, p<0.0001) from baseline to 48-week follow-up. The case series of 38 patients also reported a statistically highly significant improvement in cheek thickness (measured with ultrasound) from a pretreatment mean of 3.7 mm to 13.3 mm (p<0.0001) at a mean follow-up of 5 years.
2.3.5 The Specialist Advisers listed key efficacy outcomes as restoration of appearance, volume augmentation and psychometric evaluation of satisfaction.
2.4.1 Infection (confirmed by culture) was reported in 16% (5/32) of patients in the cohort study of 32 patients at 4-year follow-up; the median interval between NAGP injection and occurrence of infection was 2.8 years. An additional 9% (3/32) of patients had possible infection with a median time of occurrence of 3.7 years after treatment. All patients in this study with confirmed infections and 1 patient with possible infection had infections after dental treatment and were treated with antibiotics followed by surgical removal of the NAGP. The other 2 patients with possible infection were treated by antibiotic treatment alone.
2.4.2 Infections occurred in 19% (56/267) of patients in a case series of 267 patients with a median follow-up of 30 months. The rate of primary infection was 0.07 per patient-year of follow-up, and the rate of definite infection was 0.02 per patient-year of follow-up. The median time from first treatment to infection was 32 months. Surgical management with antibiotics was needed for most patients. Prior facial manipulation for dental work or cosmetic surgery near the filler site in the month before infection was the most common factor associated with infection.
2.4.3 Infections after revision procedures for correction of asymmetry were reported in 22% (4/18) of patients in a case series of 18 patients with follow-up of 2 months to 3 years. These occurred near the site of implant and were treated by antibiotics in 3 patients and by surgical removal of NAGP in 1 patient.
2.4.4 Nodules were reported in 25% (8/32) of patients in the cohort study of 32 patients at 4 years. Non-visible nodules and indurations were found in 19% (28/145) and 6% (9/145) of patients respectively in the case series of 145 patients at a mean follow-up of 50 months after NAGP injection (no further details reported).
2.4.5 Migration of the NAGP was reported in 25% (9/36) of patients at 7-year follow-up in a retrospective case series of 69 patients; mean time of onset was 12 months. This was treated by removal of the implant in 22% (8/36) of patients. Intra-oral extrusion of the NAGP through the buccal mucosa of the cheek was reported in 1 patient in the case series of 18 patients at 12-month follow-up. This was removed surgically by stab incisions and curettage.
2.4.6 The Specialist Advisers listed theoretical adverse events as vascular occlusion, delayed granuloma formation, unsatisfactory cosmetic appearance and short duration of effect.
2.5.1 The Committee recognised that facial lipoatrophy can be a very distressing condition for patients with HIV, whose quality of life may be significantly improved by effective treatment. The balance of risks and benefits of deep dermal injection of NAGP in immunocompromised patients with HIV may differ from that in patients considering the procedure for other indications.
2.5.2 The Committee noted that different types of gels used for this procedure may have different long-term effects.
2.5.3 The Committee recognised that there have been concerns about complications in the longer term after deep dermal injection of NAGP. This underpinned the recommendation for continued data collection and research, in particular to address the balance of risks and benefits in patients with HIV-related facial lipoatrophy.
2.5.4 The Committee noted reports of infection of implanted NAGP associated with subsequent dental procedures.